In this 5-year proposal, skeletal muscle ventricles (SMVs) or pumping chambers will be constructed from canine latissimus dorsi muscle and used as separate blood pumps in the circulation as either left- or right-sided cardiac assist devices after heart failure has been induced. The major objective is to determine how effective these muscle pumps are as assist devices during chronic heart failure. We have recently demonstrated that canine SMVs can pump blood effectively in the circulation for more than one year. We have also recently shown, in acute mock circulation experiments, that enough power can be obtained from canine SMVs at physiologic preloads to substitute for total left ventricular function. We plan to study which autogenous linings of SMVs will be most thrombo-resistant. The most thrombo-resistant lining will be used in SMVs to pump blood chronically after heart failure has been induced. A subgroup of SMVs pumping during chronic heart failure will be evaluated using Phosphorous-31 NMR. The purpose is to better understand the bioenergetic components of the skeletal muscle pumps which will lead to better utilization of these pumps (stimulation patterns and pumping rates). New designs in SMV pumps will be continually tested, using variations in SMV cavity size, amount of stretch placed on SMVs as they are being constructed, variations in muscle wrap and variable lengths of vascular delay and types of electrical pre-conditioning. These prototype pumps will be tested using a mock-circulation device. The most efficacious SMVs will then be used in circulation as this overall research plan progresses. Although dogs are the most frequently used animal model for skeletal muscle - cardiac assist, sheep and goats are also commonly used. We have used all three species, but have most experience with the dog and plan to use it for the majority of the experiments. However one set of experiments will compare blood flow and strength per gram of tissue at rest and during exercise of the latissimus dorsi in situ in the three different species. Blood flow and power output per gram of tissue will subsequently be evaluated in SMVs in the three different species, using a mock-circulation device. This set of experiments should help us better understand similarities and differences in the capabilities of the latissimus dorsi muscle in the three species. This research plan should yield a thorough understanding of the capabilities of SMVs as left-and-right-sided cardiac assist pumps in a chronic heart failure model. The results should contribute significantly toward potential human use of skeletal muscle ventricle-cardiac assist.